Prompting & Inference
Few-Shot Learning
Guiding a model by including a handful of examples in the prompt so it can follow the pattern.
Few-shot prompting gives the model a few demonstrations of the task inside the prompt, so it can infer the pattern and apply it to your real input. It often sharply improves results on formatting, tone, and tricky tasks without any retraining. It works because of in-context learning, the models ability to pick up a pattern on the fly from what is in its context window.
Reference: Wikipedia: Few-Shot Learning
Related terms
Zero-Shot Learning
Asking a model to do a task with no examples, relying only on instructions and what it already knows.
In-Context Learning
A model’s ability to learn a task from examples placed in the prompt, without any change to its weights.
Prompt Engineering
The practice of wording and structuring prompts to get better, more reliable results from a model.